JP5630030B2 - Rotational speed detection device for vehicle - Google Patents

Description

  The present invention relates to a vehicle rotation speed detection device that detects a rotation speed of a rotating body based on a signal interval of an interrupt signal generated by a rotation sensor for each predetermined rotation angle of the rotating body.

  2. Description of the Related Art A vehicle rotation speed detection device that includes a rotation sensor that generates an interrupt signal for each predetermined rotation angle of a rotating body and detects the rotation speed of the rotating body based on the signal interval of the interrupt signal is well known. For example, in Patent Document 1, in a wheel speed calculation device that calculates a wheel speed based on an edge of a pulse signal output from a wheel speed sensor, the edge is not detected while reflecting a change in an actual wheel speed. It is described to estimate the wheel speed during a period.

JP 2009-222601 A JP-A-6-58396

  Here, the detected rotation speed of the rotating body is used, for example, for control of each device of the vehicle, and it is desired to accurately capture the response without a response delay. For example, in shift control of an automatic transmission for a vehicle, it is important to accurately capture a change in a predetermined rotating member in the automatic transmission without delay. By the way, when calculating the rotation speed at the signal interval of the interrupt signal accompanying rotation, calculating the latest rotation speed (actual rotation speed, current rotation speed) for each latest signal interval makes it easy to ride noise ( The influence of noise increases), and the detection accuracy may deteriorate. On the other hand, in order to reduce the influence of noise, use some kind of smoothing process, such as calculating the moving average of rotation speed (that is, average rotation speed) using several consecutive signal intervals (multiple minutes). Has been proposed. However, in such an annealing process, the change in the signal interval due to the change in the actual rotation speed is similarly smoothed, so that the detection of the change in the rotation speed is delayed (that is, the response is delayed). )there is a possibility. For example, considering the case where the average rotation speed is used when calculating the constant rotation speed corresponding to the latest signal interval, this average rotation speed is obtained by calculating the average rotation speed at the intermediate point in time of a plurality of signal intervals. However, a detection response delay always occurs as compared with the current actual rotational speed. In other words, the average rotational speed should be calculated by using, for example, n signal intervals around the present time, but there is no vibration interval data at the present time. The average (current) average rotation speed cannot be obtained. Accordingly, the average rotational speed must be calculated from the data of the signal interval for the past 2n from the present time. However, since this average rotational speed is the average rotational speed for the past n times, the rotating body Unless the rotation is constant, a detection time lag, that is, a detection response delay always occurs. In particular, when the rotating body is changing at a low rotation speed, the signal interval is longer as the rotation speed is lower, and the time difference from the actual rotation speed is likely to appear, so that it appears remarkably.

  Thus, when calculating the rotation speed of the rotating body using the signal interval accompanying rotation, for example, if the rotation speed is calculated at the latest signal interval, detection response delay is unlikely to occur, but detection accuracy is improved. It tends to decrease. On the other hand, for example, if the rotation speed is calculated based on the average rotation speed using a plurality of continuous signal intervals in order to suppress a decrease in detection accuracy, a detection response delay occurs. For this reason, it is desired to improve both detection accuracy and detection response. Such a problem is not known, and it has not yet been proposed to calculate the rotation speed so as to achieve both improvement in detection accuracy and improvement in detection response.

  The present invention has been made against the background of the above circumstances. The object of the present invention is to improve both detection accuracy and detection response when detecting the rotational speed of a rotating body. It is providing the rotational speed detection apparatus for vehicles.

To achieve the above object, the gist of the present invention is: (a) a rotation sensor that generates an interrupt signal for each predetermined rotation angle of the rotating body, and the rotating body based on the signal interval of the interrupt signal; (B) calculating two types of average rotational speeds having different numbers of history of the signal intervals, and calculating the two types of average rotational speeds as actual rotational speeds. (C) The two types of average rotational speeds are calculated for different numbers of past plurals including the latest signal interval. (D) an average rotational speed calculated based on the signal interval, and (d) each time direction that occurs between the time when the estimated value of the actual rotational speed is calculated and the time corresponding to the two types of average rotational speed. Two kinds of average rotation based on delay The estimated value of the actual rotational speed is calculated as an external dividing point that divides the speed .

  In this way, two types of average rotation speeds having different numbers of signal interval histories are calculated, and the two types of average rotation speeds are placed on the same time axis as the actual rotation speed, so that the actual rotation speed is calculated. Since the estimated value of the speed is calculated, for example, an average rotational speed having a better detection accuracy than the estimated value of the actual rotational speed calculated from the latest one signal interval is used. The estimated value of the rotation speed is also relatively good in detection accuracy. In addition, since the average rotational speed at the time delayed in the time direction is calculated on the same time axis as the actual rotational speed, the estimated value of the actual rotational speed that is finally calculated is the actual rotational speed at the present time. Compared to speed, there is no response delay. Therefore, at the time of detecting the rotation speed of the rotating body, both improvement in detection accuracy and improvement in detection response can be achieved.

  Thereby, it becomes easy to grasp a physical phenomenon correctly. For example, it is possible to make the change in the rotational speed detected by the rotating body closer to the actual change. In addition, when comparing the rotational speeds of two or more different rotating bodies, for example, for detecting the operating state of an apparatus such as detecting the shift state of an automatic transmission, the detection delay of each rotational speed is detected. No need to think about. Furthermore, it is possible to advance the detection of the rotational speed of the rotating body, and it is possible to speed up the control starting point in the control operation of each device based on the detected rotational speed of each part. For example, inertia phase determination in the shifting process of an automatic transmission, control for varying the phase of the camshaft according to the state of the engine, and neutral control for disconnecting the connection between the engine and the power transmission device (eg, transaxle, transmission) when the vehicle is stopped It is possible to accurately execute control executed at a relatively low rotational speed, such as a lock-up slip control at the time of starting, in which the lock-up clutch is slipped from the start, in accordance with a change in the actual rotational speed. .

Further , the two types of average rotation speeds are average rotation speeds calculated based on different numbers of signal intervals including the latest signal interval, and the calculation of the estimated value of the actual rotation speed is calculated. Estimated value of the actual rotational speed as an external dividing point that divides the two average rotational speeds based on the respective time-direction delays occurring between the time point and the time points corresponding to the two types of average rotational speeds Therefore , for example, the estimated value of the actual rotational speed is appropriately calculated by placing two types of average rotational speeds on the same time axis as the actual rotational speed. In other words, since the two types of average rotational speed are both values that are delayed from the actual rotational speed, a value that takes into account the delay time is calculated as an estimated value of the actual rotational speed, so that detection accuracy can be improved and detected. An estimated value of the actual rotational speed that can achieve both improvement in responsiveness is calculated.

Here , preferably, the coefficient of the external component based on the delay in each time direction for calculating the estimated value of the actual rotational speed is obtained by fixing the change speed of the predetermined rotational speed to a constant value. A value that changes in accordance with the predetermined rotation speed is used. Here, when the estimated value of the actual rotational speed is calculated from the external part, it is assumed that the rotational speed change speed is originally calculated and that the speed of change is constant, so the rotational speed change speed is constant. Otherwise, an error occurs between the actual value and the estimated value by the amount of change in the rotational speed. In addition, the coefficient of the external component can be changed using the change speed of the predetermined rotation speed as a parameter, and the variation due to the parameter increases as the predetermined rotation speed decreases. However, if the original predetermined rotational speed is low, the absolute value of the estimated value of the actual rotational speed calculated by using the external component coefficient is small. As a result, if the predetermined rotational speed is low, the external component coefficient Even if the variation is large, the error as the estimated value of the actual rotational speed is reduced. Therefore, even if the change speed of the predetermined rotation speed is fixed to a constant value, the error as the estimated value of the actual rotation speed is reduced. From such a point of view, the change rate of the predetermined rotation speed is fixed to a constant value, and the coefficient of the external part that changes according to the predetermined rotation speed is used. When calculating, it is not necessary to calculate the change speed of the predetermined rotation speed, and even if the change speed of the predetermined rotation speed fixed to a constant value is used, the error as the estimated value of the actual rotation speed is reduced. Although the calculation of the change speed of the predetermined rotation speed is likely to cause an error, the present invention does not need to use the change speed of the predetermined rotation speed, and has an advantage that the error due to the change speed is originally small.

  Preferably, the predetermined rotation speed is one of an estimated value of an actual rotation speed calculated last time and the two types of average rotation speeds. If it does in this way, the coefficient of the outside part will be set up appropriately, for example.

  Preferably, the predetermined rotation speed is an average rotation speed that is temporally closer to the estimated value of the actual rotation speed among the two types of average rotation speeds. In this way, for example, the average rotation speed with a relatively small response delay is used, and the error as the estimated value of the actual rotation speed is relatively small. In addition, if the estimated value of the actual rotational speed to be calculated is used, the calculation load becomes recursive and the calculation load increases. However, if the average rotational speed is used, such a recursive calculation can be avoided.

  Further, preferably, the estimated value of the actual rotational speed is limited to be a numerical value in a range that can be taken as the actual rotational speed. In this way, for example, it is avoided that the estimated value of the actual rotational speed is a numerical value that is physically impossible.

Further, another gist of the invention for achieving the above object is that: (a) a rotation sensor that generates an interrupt signal for each predetermined rotation angle of the rotating body is provided, and based on the signal interval of the interrupt signal; An apparatus for detecting a rotational speed of a vehicle for detecting the rotational speed of the rotating body, wherein (b) an estimated value of an actual rotational speed based on one rotational speed calculated in the past and a change speed of the one rotational speed (C) From the preset relationship between the one rotational speed and the estimated value of the actual rotational speed using the change speed of the one rotational speed as a parameter, the one rotational speed and An estimated value of the actual rotational speed is calculated based on the change speed of the one rotational speed .

  In this way, since the estimated value of the actual rotational speed is calculated based on one rotational speed calculated in the past and the change speed of the one rotational speed, for example, it is calculated from the latest one signal interval. The average rotational speed with better detection accuracy compared to the actual rotational speed estimate and the actual rotational speed estimate calculated last time with good detection accuracy calculated using the average rotational speed are used. The estimated value of the actual rotational speed calculated in (1) also has a relatively good detection accuracy. In addition, since the rotational speed at the time delayed in the time direction is calculated as the rotational speed at the same time as the actual rotational speed, the estimated value of the actual rotational speed that is finally calculated is the actual rotational speed at the present time. Compared to speed, there is no response delay. Therefore, at the time of detecting the rotation speed of the rotating body, both improvement in detection accuracy and improvement in detection response can be achieved. Further, the present invention has an advantage that only one rotation speed calculated in the past is required.

  Thereby, it becomes easy to grasp a physical phenomenon correctly. For example, it is possible to make the change in the rotational speed detected by the rotating body closer to the actual change. In addition, when comparing the rotational speeds of two or more different rotating bodies, for example, for detecting the operating state of an apparatus such as detecting the shift state of an automatic transmission, the detection delay of each rotational speed is detected. No need to think about. Furthermore, it is possible to advance the detection of the rotational speed of the rotating body, and it is possible to speed up the control starting point in the control operation of each device based on the detected rotational speed of each part. For example, inertia phase determination in the shifting process of an automatic transmission, control for varying the phase of the camshaft according to the state of the engine, and neutral control for disconnecting the connection between the engine and the power transmission device (eg, transaxle, transmission) when the vehicle is stopped It is possible to accurately execute control executed at a relatively low rotational speed, such as a lock-up slip control at the time of starting, in which the lock-up clutch is slipped from the start, in accordance with a change in the actual rotational speed. .

Also, from a preset relationship between the estimated value of that one of the rotational speed variation rate of the previous SL one rotational speed as parameters the actual rotational speed, changes in the one rotational speed and that one of the rotational speed since to calculate the actual estimate of the rotational speed based on the speed, for example, the estimated value of the actual rotational speed based on a single rotation speed calculated in the past and its one rotational speed of the speed of change Calculated appropriately. In other words, since one rotation speed calculated in the past is a value delayed from the actual rotation speed, a value considering the delay time is calculated as an estimated value of the actual rotation speed. An estimated value of the actual rotational speed that can achieve both improvement in detection responsiveness is calculated.

  Preferably, the preset relationship is a direct relationship between the one rotational speed and the estimated value of the actual rotational speed. In this way, for example, the estimated value of the actual rotation speed is more reliably calculated based on one rotation speed calculated in the past and the change speed of the one rotation speed.

  Preferably, the preset relationship is a relationship between the one rotation speed and a difference between the one rotation speed and an estimated value of the actual rotation speed. In this way, for example, the estimated value of the actual rotation speed is more reliably calculated based on one rotation speed calculated in the past and the change speed of the one rotation speed.

  Preferably, the preset relationship is a relationship between the one rotation speed and a rotation ratio between the one rotation speed and the estimated value of the actual rotation speed. In this way, for example, the estimated value of the actual rotation speed is more reliably calculated based on one rotation speed calculated in the past and the change speed of the one rotation speed.

  Preferably, the one rotation speed is an estimated value of an actual rotation speed calculated last time or an average rotation speed calculated based on the signal intervals for a plurality of past times. In this way, for example, an estimated value of the actual rotational speed is appropriately calculated.

  Further, preferably, the estimated value of the actual rotational speed is limited to be a numerical value in a range that can be taken as the actual rotational speed. In this way, for example, it is avoided that the estimated value of the actual rotational speed is a numerical value that is physically impossible.

  Preferably, the rotation sensor is provided at a position facing a plurality of detected objects arranged in a circumferential direction on a radially outer peripheral portion of the rotating body, and accompanying the rotation of the rotating body. An interrupt signal is generated on the basis of sequentially facing the detected object. In this way, for example, the interrupt signal is appropriately generated.

  Preferably, the vehicle including the vehicle rotational speed detection device includes a vehicle power transmission device in a power transmission path from a power source to the drive wheels. As this power source, for example, a gasoline engine such as an internal combustion engine that generates power by combustion of fuel, a diesel engine, or the like is preferably used. However, other prime movers such as an electric motor may be employed alone or in combination with the engine. it can.

  Preferably, the vehicle power transmission device includes a transmission mechanism unit alone, a torque converter, a transmission mechanism unit having a plurality of transmission ratios, or a speed reduction mechanism unit and a differential mechanism unit in addition to the transmission mechanism unit. The In this speed change mechanism portion, a plurality of gear stages (shift speeds) are alternatively achieved by selectively connecting the rotating elements of a plurality of sets of planetary gear devices by an engagement device, for example, forward four speeds, forward Various planetary gear type automatic transmissions having 5 speeds, 6 forward speeds, and more, etc., and a plurality of pairs of transmission gears that always mesh with each other between two shafts, and any of these multiple pairs of transmission gears Is a synchronous mesh type parallel twin-shaft transmission that is alternatively in a power transmission state by a synchronizer, and the gear stage is automatically controlled by a synchronizer driven by a hydraulic actuator, although it is a synchronous mesh type parallel twin-shaft transmission. A synchronous mesh type parallel two-shaft automatic transmission and a synchronous mesh type parallel two-shaft automatic transmission, which have two input shafts and a clutch is connected to the input shaft of each system, respectively. Even number and odd number The so-called DCT (Dual Clutch Transmission) which is a type of transmission connected to the power transmission belt, a transmission belt functioning as a power transmission member is wound around a pair of variable pulleys with variable effective diameters, and the transmission ratio is continuously variable An automatic transmission that is a so-called belt-type continuously variable transmission, a pair of cones that are rotated around a common axis, and a plurality of rollers that can rotate around the axis and that can rotate about a center of rotation. The automatic transmission is a so-called traction type continuously variable transmission in which the transmission ratio is variable by changing the crossing angle between the rotation center of the roller and the shaft center. 1 A differential mechanism constituted by, for example, a planetary gear device that distributes to an electric motor and an output shaft, and a second electric motor provided on the output shaft of the differential mechanism. The gear ratio is electrically transmitted by mechanically transmitting the main part of the power from the engine to the drive wheel side and electrically transmitting the remaining part of the power from the engine using the electric path from the first motor to the second motor. It is composed of an automatic transmission that functions as an electric continuously variable transmission that is changed, or an automatic transmission that is mounted on a so-called parallel hybrid vehicle equipped with an electric motor capable of transmitting power to an engine shaft, an output shaft, etc. The

  Preferably, the rotating body may be fixed to a rotating member that needs to detect a rotation speed and rotate together with the rotating member, or the rotating member itself. For example, the rotating body includes an input rotating member and an output rotating member of the speed change mechanism unit, an output rotating member (crankshaft) of the engine, a rotating member that rotates together with a wheel, or a rotating member that is fixed to each rotating member together with the rotating member. Each is a rotating rotor or drum.

It is a figure explaining schematic structure of the power transmission path which constitutes vehicles to which the present invention is applied, and explains the principal part of a control system provided in vehicles for detecting (calculating) the rotational speed of output gears and the like. It is a block diagram to do. It is the schematic explaining an example of each sensor for detecting the rotational speed of each rotation members, such as a crankshaft of an engine, the input shaft of a transmission mechanism part, an output gear, and each wheel. It is a figure explaining an example which calculates the newest rotational speed in the rotary body of FIG. It is a figure which shows the coefficient of an external part when calculating an estimated rotational speed using 1/4 rotational average rotational speed and 1 rotational average rotational speed, Comprising: The relationship between the estimated rotational speed calculated last time and the coefficient of external part Is illustrated. It is a figure which shows the coefficient of an external part when calculating an estimated rotational speed using 1/4 rotational average rotational speed and 1 rotational average rotational speed, Comprising: It illustrates the relationship. FIG. 4 shows the error for each inclination when the estimated rotation speed is calculated by applying the coefficient value when the inclination is 600 [rpm / sec] to the case where the inclination is 0,300,600,900,1200 [rpm / sec]. FIG. FIG. 5 shows the error for each inclination when the estimated rotation speed is calculated by applying the coefficient value when the inclination is 600 [rpm / sec] to the case where the inclination is 0,300,600,900,1200 [rpm / sec]. FIG. It is a functional block diagram explaining the principal part of the control function by the electronic controller of FIG. It is a flowchart explaining the control action for making the improvement of detection accuracy and the improvement of detection responsiveness compatible at the time of the detection of the principal part of the control action of an electronic controller, ie, the rotational speed of a rotary body. It is a time chart which shows an example at the time of performing the control action shown to the flowchart of FIG. It is an estimated rotation speed map that is set in advance between the 1/4 rotation average rotation speed and the estimated rotation speed itself with the change speed of the 1/4 rotation average rotation speed as a parameter. FIG. 5 is an estimated rotation speed difference map set in advance between a 1/4 rotation average rotation speed and a difference between the 1/4 rotation average rotation speed and the estimated rotation speed with a change speed of the 1/4 rotation average rotation speed as a parameter; A 1/4 estimated rotation speed map with 1/4 rotation average rotation speed and 1/4 rotation average rotation speed to estimated rotation speed as a parameter. is there. FIG. 10 is a flowchart for explaining a control operation for achieving both improvement in detection accuracy and improvement in detection response when detecting the main part of the control operation of the electronic control unit, that is, the rotational speed of the rotating body, corresponding to the flowchart of FIG. 9. This is another embodiment.

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

  FIG. 1 is a diagram illustrating a schematic configuration of a power transmission path from an engine 12 to a drive wheel 14 constituting a vehicle 10 to which the present invention is applied, and the rotational speed of an output gear 18 and the like of a transmission mechanism unit 16. FIG. 2 is a block diagram illustrating a main part of a control system provided in vehicle 10 for detection (calculation). In FIG. 1, the transmission mechanism 16 is preferably used in, for example, an FF (front engine / front drive) type vehicle that is placed horizontally in a vehicle, and is an engine 12 that is an internal combustion engine as a driving force source for traveling. From the output gear 18 as an output rotating member constituting one of the counter gear pairs 20 to the power, the counter gear pair 20 as a power transmission device, the final gear pair 22, the differential gear device (differential gear) 24, and a pair of axles (Drive shaft (D / S)) 26 and the like are sequentially transmitted to the pair of drive wheels 14. The transmission mechanism 16, the counter gear pair 20, the final gear pair 22, the differential gear device (differential gear) 24, and the like constitute a transaxle (T / A).

  Further, the vehicle 10 is provided with an electronic control unit 80 including a part of a vehicle rotation speed detection device for detecting (calculating) the rotation speed of each rotation member in the power transmission path, for example. The electronic control unit 80 includes, for example, a so-called microcomputer having a CPU, a RAM, a ROM, an input / output interface, and the like, and the CPU uses a temporary storage function of the RAM according to a program stored in the ROM in advance. By performing signal processing, output control of the engine 12 and shift control of the transmission mechanism unit 16 are executed.

The electronic control unit 80, for example the crankshaft of the crank angle of the engine 12 from the engine rotational speed sensor 28 (position) A _CR and the engine rotational speed signal corresponding to the engine rotational speed N _E, the shift from the input rotation speed sensor 30 The input rotational speed signal corresponding to the rotational speed of the input shaft as the input rotational speed N _IN of the mechanism unit 16, the rotational speed of the output gear 18 as the output rotational speed N _OUT of the transmission mechanism unit 16 from the output rotational speed sensor 32. A corresponding output rotation speed signal, a wheel speed signal corresponding to the rotation speed N _W of each wheel (that is, each wheel including the driven wheel 14) from each wheel speed sensor 34 are supplied. In the vehicle 10 in which power from the engine 12 is transmitted to the transmission mechanism 16 side via a known torque converter or the like, the input rotational speed N _IN of the transmission mechanism 16 described above is, for example, a turbine of a torque converter. This corresponds to the rotational speed _NT . The output rotational speed N _{OUT of the} speed change mechanism 16 and the rotational speed N _W of each wheel correspond to the vehicle speed V, for example.

  The electronic control device 80 outputs, for example, an engine output control command signal for output control of the engine 12, a shift control command signal for shift control of the transmission mechanism unit 16, and the like. For example, the electronic control unit 80 outputs an engine output control command signal to a throttle actuator, a fuel injection device, an igniter, etc. (not shown) based on the engine rotation speed signal, the shift control command signal, etc., and controls the output of the engine 12. Run. Further, the electronic control unit 80 outputs a shift control command signal to a hydraulic control circuit (not shown) or the like based on the input rotation speed signal, the output rotation speed signal, etc., and executes the gear ratio switching control of the transmission mechanism unit 16. To do.

  2 shows an engine rotation speed sensor 28 for detecting a rotation speed N of each rotary member 50 such as a crankshaft of the engine 12, an input shaft of the transmission mechanism section 16, an output gear 18 of the transmission mechanism section 16, and each wheel. It is the schematic explaining an example of the sensor 60 as each rotation sensor, such as the input rotation speed sensor 30, the output rotation speed sensor 32, and the wheel speed sensor. In FIG. 2, the rotating body 70 is, for example, a rotation detecting rotor or a rotation detecting drum that is integrally fixed with the rotating member 50 on the same axis C and rotates together with the rotating member 50. Further, the sensor 60 is provided at a position facing a plurality of teeth 72 as detection bodies arranged along the circumferential direction on the radially outer peripheral portion of the rotating body 70, for example, and the rotational speed of the rotating member 50. This constitutes a part of a vehicle rotational speed detection device for detecting N. As the sensor 60, for example, an electromagnetic pickup type, Hall element type, MRE (Magnetic Resistance Element) type sensor or the like is employed.

  For example, in the case of an electromagnetic pickup sensor, the rotation of the rotating body 70 causes the air gap between the sensor 60 and the rotating body 70 to be alternately opposed to the teeth 72 and the recesses 74. An alternating voltage whose frequency changes corresponding to the rotation speed of the sensor 60 is generated from the sensor 60. This AC voltage is supplied to the electronic control device 80 as a rotation speed signal, and the AC voltage supplied by the electronic control device 80 is converted into a rectangular wave pulse signal P as shown in FIGS. Converted. Further, for example, in the case of a Hall element type or MRE type sensor, the sensor is composed of a sensor circuit (IC) incorporating a Hall element or MRE. When the rotating body 70 rotates, a magnetic field applied to the sensor circuit (IC) is changed. Therefore, an alternating voltage whose frequency changes in accordance with the rotational speed of the rotating body 70 is generated, and the alternating voltage is converted into a rectangular pulse signal P as shown in FIGS. Is done. The pulse signal P is supplied to the electronic control unit 80 as a rotation speed signal.

In the rotating body 70 shown in FIG. 2 (a), for example, 16 teeth 72 having substantially the same shape and substantially equal teeth a to teeth p are formed by stamping or the like, and as a result, 16 teeth between the teeth are formed. A recess 74, which is a recess a to a recess p, is formed as an object to be detected. In addition, the electronic control unit 80 uses the intervals of the pulse signals Pa to Pp corresponding to the teeth a to p and the recesses a to p in the pulse signal P as shown in FIGS. Each signal interval tN [i] as (pulse interval) is detected (calculated). In this embodiment, each signal interval tN [i] is a period from the rise of the pulse signal P to the rise of the next pulse signal P. That is, in this embodiment, the signal interval tN [i] is an interval between interrupt signals (hereinafter referred to as interrupt signals _SPU ) corresponding to the rising edges of the pulse signals P. In FIG. 2B, for example, each pulse signal Pa to Pp becomes high (Hi) at a position where each tooth a to tooth p is substantially opposed to the sensor 60, and each recess a to p is substantially omitted from the sensor 60. Each pulse signal Pa to Pp becomes low at the opposite position. In FIG. 2C, for example, when each tooth a to tooth p is substantially opposed to the sensor 60, each pulse signal Pa to Pp is turned on (ON) for a certain period, and then each tooth a to tooth p is turned on. The pulse signals Pa to Pp are turned off until they are substantially opposed to the sensor 60. Then, the electronic control unit 80 calculates the rotational speed N of the rotating body 70 using each signal interval tN [i]. As described above, the sensor 60 and the electronic control unit 80 generate the pulse signals Pa to Pp (interrupt signal S _PU ) based on the fact that the sensor 60 is sequentially opposed to the teeth 72 as the rotating body 70 rotates. namely generates an interrupt signal S _PU at every predetermined rotation angle of the rotating body 70 corresponding to each interval of the teeth 72 formed at substantially equal intervals, each signal interval tN [i of each interrupt signal S _PU ], The rotational speed N of the rotating body 70 is detected. The signal interval tN [i] represents a signal interval i times before the current time. For example, if i = 0, the current (current) signal interval is indicated. If i = 1, the previous time ( It represents the signal interval of the previous time. That is, for example, the signal interval tN [0] represents the latest (current) signal interval, and the signal interval tN [1] represents the previous signal interval.

  Here, when the signal interval tN [i] accompanying the rotation of the rotating body 70 is converted into the rotation speed N, the signal interval tN [i] for each tooth of the teeth a to p (that is, the latest one signal interval tN). In the case of directly converting [i]) to the rotational speed N, the detection response is excellent, but on the other hand, for example, noise is applied to a short signal interval tN [i] of one tooth or each tooth There is a possibility that the detection accuracy deteriorates due to the processing variation in 72 and each recess 74. In contrast, when the rotational speed N is calculated by moving average using a plurality of continuous signal intervals tN [i], the detection accuracy is excellent, but on the other hand, the continuous multiple signal intervals tN. Only the average rotational speed at the time intermediate point of [i] is obtained, and a detection response delay always occurs as compared with the current (current) actual rotational speed N (actual rotational speed N).

  In view of this, the electronic control unit 80 according to the present embodiment, when detecting the rotational speed N of the rotating body 70, for example, in order to achieve both improvement in detection accuracy and improvement in detection response, for example, the signal interval tN [ In order to achieve both detection responsiveness when i] is directly converted to rotation speed N and detection accuracy when calculating average rotation speed using a plurality of consecutive signal intervals tN [i], Two types of average rotation speeds having different numbers of histories at the interval tN [i] are calculated, and the two types of average rotation speeds are placed on the same time axis as the actual rotation speed N, so that an estimated value of the actual rotation speed N (hereinafter referred to as the actual rotation speed N). , Estimated rotational speed N0).

FIG. 3 is a diagram for explaining an example of calculating the estimated rotational speed N0 as the latest rotational speed corresponding to the actual rotational speed N in the rotating body 70 of FIG. In FIG. 3, for example, signal intervals tN [i] corresponding to the pulse signals Pa to Pe generated corresponding to the teeth a to p are sequentially calculated. For example, the signal interval tN [0] and the signal interval tN [ 1], signal intervals tN [2]... Are stored in the RAM of the electronic control unit 80. In this embodiment, since the rotational speed N decreases with the passage of time, that is, the slope corresponding to the change speed ΔN of the rotational speed N is a downward slope, the signal interval tN [i] is increased with the passage of time. Have been long. In this embodiment, as the rotation speed N, for example, the turbine rotation speed N _T (input rotation speed N _IN ) is exemplified. Further, the change speed ΔN of the rotation speed N corresponds to the change amount of the rotation speed N, and agrees with the change amount in the repeated control.

Then, the electronic control unit 80, as one of two types of average rotation speeds having different numbers of histories of the signal interval tN [i], for example, an average rotation speed having four histories of the signal interval tN [i], that is, the latest signal. The 1/4 rotation average rotational speed N4 calculated based on the signal interval (tN [0] + tN [1] + tN [2] + tN [3]) for the past four including the interval tN [0] is expressed by the following formula ( Calculate according to 1). In addition, the electronic control unit 80, as the other of two types of average rotation speeds having different numbers of histories of the signal interval tN [i], for example, an average rotation speed of 16 histories of the signal interval tN [i], that is, the latest signal. One rotation average rotational speed N16 calculated based on the signal interval (tN [0] + tN [1] +... + TN [15]) for the past 16 including the interval tN [0] is calculated according to the following equation (2). To do. The constant K in the following formulas (1) and (2) is obtained from the unit signal interval tN [i] determined according to the unit of the rotational speed N such as [rpm] or [rad / sec]. It is a predetermined constant for the purpose.
N4 = K / ((tN [0] + tN [1] + tN [2] + tN [3]) / 4) (1)
N16 = K / ((tN [0] + tN [1] +... + TN [15]) / 16) (2)

  As is clear from FIG. 3, the 1/4 rotation average rotation speed N4 is the average rotation speed at the time halfway point during the most recent 1/4 rotation, that is, from the current point ((tN [0] + tN [ 1] + tN [2] + tN [3]) / 2) Average rotational speed at time point b ′ before. Further, the average rotation speed N16 per rotation is the average rotation speed at the time intermediate point during the latest one rotation, that is, ((tN [0] + tN [1] +... + TN [15]) / 2) Average rotational speed at the previous time point a ′. As described above, the 1/4 rotation average rotation speed N4 and the 1 rotation average rotation speed N16 are values delayed from the present time when the actual rotation speed N is calculated. Therefore, as shown by the circles in FIG. 3, the estimated rotational speed N0 is calculated by placing these two types of average rotational speeds on the same time axis as that at which the actual rotational speed N is calculated, that is, on the same current time.

  For example, as shown in FIG. 3, a time-direction delay a that occurs between the calculation time point of the actual rotation speed N, that is, the calculation time point (current time) of the estimated rotation speed N0 and the time point a ′ corresponding to the average rotation speed N16. Between “(= (tN [0] + tN [1] +... + TN [15]) / 2) and the estimated rotation speed N0 and the time b ′ corresponding to the 1/4 rotation average rotation speed N4. Average rotation speed on the current time axis based on the time direction delay b ′ (= (tN [0] + tN [1] + tN [2] + tN [3]) / 2) An external dividing point that divides the rotational speed difference (length between each point) between N16 (point N16 corresponding to N16) and 1/4 rotation average rotational speed N4 (point N4 corresponding to N4) into a: b. Estimated rotational speed N0 is calculated as (point N0).

Specifically, the relationship shown in the following formula (3) is established on the basis of the external division, and when α = a / b (> 1) is set, the estimated rotational speed N0 is expressed by the following formula (4). . When α ′ = α / (α−1) is set, the estimated rotational speed N0 is expressed by the following formula (5). Here, if the change speed ΔN of the rotational speed N is constant, a: b = a ′: b ′, and α is expressed by the following equation (6). In the present embodiment, α and α ′ are used as external component coefficients in a mathematical expression for obtaining an external division point from the external component ratio (a: b).
(N16-N0) :( N4-N0) = a: b (3)
N0 = α / (α-1) × N4-1 / (α-1) × N16 (4)
N0 = α ′ × N4 + (1−α ′) × N16 (5)
α = (tN [0] + tN [1] +... + tN [15])
/ (tN [0] + tN [1] + tN [2] + tN [3]) (6)

  In this way, the estimated rotational speed N0 can be calculated using both the 1/4 rotational average rotational speed N4 and the 1 rotational average rotational speed N16 so as to achieve both improved detection accuracy and improved detection response. . However, the calculation of the estimated rotation speed N0 is based on the assumption that the change speed ΔN of the rotation speed N is constant. That is, it is assumed that the change speed ΔN4 at the 1/4 rotation average rotation speed N4 and the change speed ΔN16 at the 1 rotation average rotation speed N16 are both the same as the change speed ΔN at the actual rotation speed N. Therefore, if the change speed ΔN of the rotation speed N is not constant, an error E is generated between the actual rotation speed N and the estimated rotation speed N0 by the amount that the change speed ΔN has changed. Hereinafter, the relationship between the difference in the change speed ΔN of the rotational speed N and the error E will be described.

  FIGS. 4 and 5 are diagrams showing the coefficient α ′ of the external component when the estimated rotation speed N0 is calculated using the 1/4 rotation average rotation speed N4 and the 1 rotation average rotation speed N16, respectively. FIG. 4 illustrates the relationship between the estimated rotational speed N0 calculated last time and the external component coefficient α ′, and FIG. 5 illustrates the relationship between the 1/4 rotation average rotational speed N4 and the external component coefficient α ′. To do. 4 and 5, the change speed (slope) ΔN (here, ΔN0 and ΔN4) of the rotational speed N (here, the estimated rotational speed N0 and the quarter-turn average rotational speed N4 calculated previously) is lowering to the right. It is a slope. 4 and 5, the coefficient α ′ of the external part changes according to the rotational speed N, and the lower the rotational speed N, the variation of the external part coefficient α ′ with respect to the change speed (slope) ΔN of the rotational speed N. Becomes larger. However, when the rotational speed N is originally low, the absolute value of the estimated rotational speed N0 (see the above equation (5)) calculated by multiplying the coefficient α ′ becomes small. As a result, if the rotational speed N is low, Even if the variation of the coefficient α ′ is large, the error E of the estimated rotational speed N0 is small.

  6 and 7 show the values of the coefficient α ′ when the change speed (slope) ΔN is 600 [rpm / sec] in FIGS. 4 and 5, respectively, and the change speed (slope) ΔN is 0,300,600,900,1200 [rpm / FIG. 6 is a diagram showing an error E at each change speed (slope) ΔN when calculating the estimated rotational speed N0 when applied to the case of [sec]. 6 exemplifies the relationship between the previously calculated estimated rotational speed N0 and the error E corresponding to FIG. 4, and the relationship between the 1/4 rotational average rotational speed N4 and the error E corresponding to FIG. This is just an example. 6 and 7, when the change speed (slope) ΔN is 600 [rpm / sec], the error E is zero, and at other change speeds (slopes) ΔN, the lower the rotation speed N, the lower the speed E is 600 [rpm / sec. The error E is increased mainly in the case of], but the error E is suppressed to about 5 [rpm] or less in any case as an absolute value.

  Thus, when the estimated rotational speed N0 is calculated using the 1/4 rotational average rotational speed N4 and the 1 rotational average rotational speed N16, it is assumed that the change speed ΔN of the rotational speed N is essentially constant. When the change rate ΔN is not constant, an error E is generated by the amount of variation of the coefficients α and α ′ of the external component depending on the change rate ΔN. However, as described above, for example, even if the change speed ΔN of the rotational speed N is fixed to a constant value, the error E caused thereby is very small with respect to the actual rotational speed N, and is at a level that does not cause a problem. Therefore, in this embodiment, the estimated rotational speed N0 is calculated with the change speed (slope) ΔN fixed at a constant value regardless of the actual change speed (slope) ΔN. That is, as the coefficients α and α ′ of the external component, values that change according to the predetermined rotational speed N after the change speed ΔN of the predetermined rotational speed N is fixed to a constant value. For example, in FIG. 4 or FIG. 5, the change rate ΔN is fixed to a constant value of 600 [rpm / sec], and an external component coefficient α ′ that changes according to the rotation speed N is used. Then, the estimated rotational speed N0 is calculated according to the equation (5) using the coefficient α ′ of the outer part. 4 and 5 also show a relationship (coefficient map) obtained and experimentally (or calculated) in advance between the rotational speed N and the coefficient α ′. Moreover, 0,300,600,900,1200 [rpm / sec] of the change speed ΔN is an example from the range that the change speed ΔN assumed in the actual machine can take, and is a value of 600 [rpm / sec] fixed as a constant value. Corresponds to its median. As the predetermined rotation speed N, the previously calculated estimated rotation speed N0 or two types of rotation speeds (1/4 rotation average rotation speed N4, 1 rotation average rotation speed N16) may be used. However, the estimated rotation speed N0 is a value to be calculated, and if the estimated rotation speed N0 is used, there is a possibility that the calculation load increases due to recursive calculation. It is appropriate to use. Further, of the two types of average rotation speeds, the 1/4 rotation average rotation speed N4 having a response delay smaller than the one rotation average rotation speed N16 with respect to the actual rotation speed N, that is, close to the estimated rotation speed N0 in time. It is more appropriate to use the 1/4 rotation average rotation speed N4.

More specifically, FIG. 8 is a functional block diagram for explaining the main part of the control function by the electronic control unit 80. In FIG. 8, the signal interval calculation unit, that is, the signal interval calculation means 82 uses, for example, the latest signal interval tN [0] corresponding to the latest interrupt signal S _PU new generated along with the rotation of the rotating body 70. Are sequentially calculated for each interrupt signal S _PU new (for each latest tooth), and the calculation result is stored in the RAM. At this time, the previously calculated tN [0], signal interval tN [1], signal interval tN [2]... Are tN [1], signal interval tN [2], signal interval tN [3]. To be rewritten. For example, when one average rotation speed N16 and one quarter rotation average rotation speed N4 are used as two types of average rotation speeds, a signal interval tN [i] for calculating the one rotation average rotation speed N16 is used. It is sufficient that at least the history is stored.

The average rotation speed calculation unit, that is, the average rotation speed calculation means 84, for example, obtains the signal intervals (tN [0] + tN [1] + tN [2] for the past four including the latest signal interval tN [0] from the equation (1). ] + TN [3]), the 1/4 rotation average rotation speed N4 is sequentially calculated for each latest interrupt signal _SPU new. Further, the average rotation speed calculation means 84, for example, from the equation (2), the signal intervals for the past 16 (tN [0] + tN [1] +... + TN [15]) including the latest signal interval tN [0]. 1 rotation average rotation speed N16 is sequentially calculated for each latest interrupt signal _SPU new.

The estimated rotation speed calculation unit, that is, the estimated rotation speed calculation means 86 calculates the average rotation speed from a coefficient map in which the change speed ΔN in the coefficient map as shown in FIG. 5 is fixed to a constant value (for example, 600 [rpm / sec]). Based on the 1/4 rotation average rotation speed N4 calculated by the means 84, the coefficient α ′ of the external part is calculated. Next, the estimated rotational speed calculating means 86 is based on, for example, the coefficient α ′ from the equation (5) and the 1/4 rotational average rotational speed N4 and the single rotational average rotational speed N16 calculated by the average rotational speed calculating means 84. Thus, the estimated rotational speed N0 is sequentially calculated for each latest interrupt signal _SPU new.

  The estimated value guard unit, that is, the estimated value guard unit 88 limits (guards) the estimated rotation speed N0 calculated by the estimated rotation speed calculation unit 86, for example, to be a numerical value within a range that can be taken as the actual rotation speed N. For example, the estimated value guard means 88 determines whether or not the estimated rotational speed N0 calculated by the estimated rotational speed calculating means 86 is within a substantial variation range of the actual rotational speed N that is obtained experimentally in advance and set. If the estimated rotational speed N0 is not within the substantial fluctuation range of the actual rotational speed N, a command to set the value closest to the estimated rotational speed N0 within the substantial fluctuation range to the estimated rotational speed N0 is issued. Output to the calculation means 86. Specifically, when a value greater than or equal to zero is set in advance as the substantial fluctuation range of the actual rotational speed N, and the estimated rotational speed N0 calculates a negative value as the estimated rotational speed N0, The estimated rotation speed calculation means 86 outputs a command for setting the estimated rotation speed N0 to zero to the estimated rotation speed calculation means 86.

FIG. 9 is a flowchart for explaining a control operation for achieving both improvement in detection accuracy and improvement in detection response when detecting a main part of the control operation of the electronic control unit 80, that is, the rotational speed N of the rotating body 70. For example, it is repeatedly executed with an extremely short cycle time of about several milliseconds to several tens of milliseconds. FIG. 10 is a time chart when the control operation shown in the flowchart of FIG. 9 is executed when the turbine rotational speed _NT is detected (calculated).

In FIG. 9, first, in a step (hereinafter, step is omitted) S10 corresponding to the signal interval calculation means 82, for example, the latest corresponding to the latest interrupt signal _SPU new generated with the rotation of the rotating body 70, for example. The signal interval tN [0] is sequentially calculated for each latest interrupt signal S _PU new (for each latest tooth), and the calculation result is stored in the RAM. At this time, tN [0], signal interval tN [1], signal interval tN [2]... Calculated previously are tN [1], signal interval tN [2], signal interval tN [3]. To be rewritten. Next, in S20 corresponding to the average rotation speed calculation means 84, for example, the 1/4 rotation average rotation speed N4 is calculated based on the past four signal intervals including the latest signal interval tN [0] from the equation (1). It is calculated sequentially for each latest interrupt signal _SPU new. Further, for example, based on the past 16 signal intervals including the latest signal interval tN [0] from the equation (2), one rotation average rotation speed N16 is sequentially calculated for each latest interrupt signal _SPU new. . Next, in S30 corresponding to the estimated rotation speed calculation means 86, for example, the 1/4 rotation average rotation calculated in S20 from the coefficient map in which the change speed ΔN is fixed to a constant value (for example, 600 [rpm / sec]), for example. Based on the speed N4, the coefficient α ′ of the external part is calculated. Next, in S40 corresponding to the estimated rotation speed calculating means 86 and the estimated value guard means 88, for example, the coefficient α ′ calculated in the above S30 from the equation (5) and the 1/4 rotation average calculated in the above S20. Based on the rotation speed N4 and the one-rotation average rotation speed N16, the estimated rotation speed N0 is sequentially calculated for each latest interrupt signal _SPU new. Note that the estimated rotational speed N0 calculated here is limited (guarded) so as to be a numerical value in a physically possible range as the actual rotational speed N.

Thus, as shown in FIG. 10, when the turbine rotational speed N _T (input rotational speed N _IN ) is changed, 1/4 rotational average rotational speed N4 (broken line) and 1 calculated by simple moving average Compared with the rotation average rotation speed N16 (two-dot chain line), the estimated rotation speed N0 (solid line) with high responsiveness and accuracy is calculated. In particular, the detection accuracy in the low rotation speed region is improved, and the control using this low rotation speed region is improved.

  As described above, according to the present embodiment, two types of average rotation speeds (for example, 1/4 rotation average rotation speed N4 and 1 rotation average rotation speed N16) having different numbers of histories of the signal interval tN [i] are calculated. Since the estimated rotational speed N0 is calculated by placing the two types of average rotational speeds on the same time axis as the actual rotational speed N, for example, the estimated rotational speed calculated from the signal interval tN [i] of the latest single tooth Since an average rotation speed with better detection accuracy than N0 is used, the estimated rotation speed N0 finally calculated also has a relatively good detection accuracy. In addition, since the average rotational speed at the time delayed in the time direction is calculated on the same time axis as the actual rotational speed N, the finally calculated estimated rotational speed N0 is compared with the current actual rotational speed N. Thus, there is no response delay. Therefore, when detecting the rotational speed N of the rotating body 70, it is possible to achieve both improvement in detection accuracy and improvement in detection response.

  Thereby, it becomes easy to grasp a physical phenomenon correctly. For example, the change in the rotational speed N detected by the rotating body 70 can be made closer to the actual change. Further, when comparing the rotational speeds of two or more different rotating bodies, for example, for detecting the operating state of the in-vehicle device such as detecting the speed change state of the speed change mechanism unit 16, There is no need to consider detection delay. Further, it is possible to advance the detection of the rotational speed N of the rotating body 70, and it is possible to accelerate the control starting point in the control operation of each device based on the detected rotational speed of each part. For example, inertia phase determination in the shifting process of the transmission mechanism 16, control to vary the phase of the camshaft according to the state of the engine 12, and coupling between the engine 12 and a power transmission device (for example, transaxle, transmission) when the vehicle is stopped Controls executed at a relatively low rotational speed, such as a neutral control for separating and a lock-up slip control for starting to lock the lock-up clutch from the start, can be accurately executed in accordance with changes in the actual rotational speed N. It becomes possible.

  Further, according to the present embodiment, the two types of average rotation speeds are calculated based on different numbers of past signal intervals tN [i] including the latest signal interval tN [0]. An external division that divides the two average rotation speeds based on the delays in the respective time directions that occur between the calculation time point of the estimated rotation speed N0 and the time points corresponding to the two average rotation speeds. Since the estimated rotational speed N0 is calculated as a point, the estimated rotational speed N0 is appropriately calculated by placing two types of average rotational speeds on the same time axis as the actual rotational speed N, for example. That is, since the two types of average rotation speeds are values that are delayed from the actual rotation speed N, a value that takes the delay time into account is calculated as the estimated rotation speed N0, so that the detection accuracy can be improved and the detection responsiveness can be improved. An estimated rotation speed N0 that can achieve both improvement is calculated.

  In addition, according to the present embodiment, the coefficients α and α ′ of the external component based on the respective time-direction delays for calculating the estimated rotational speed N0 make the change speed ΔN of the predetermined rotational speed N constant. Since a value that changes according to the predetermined rotational speed N is used after being fixed, it is not necessary to calculate the change speed ΔN of the predetermined rotational speed N when calculating the estimated rotational speed N0, and to a constant value. Even if the fixed change speed ΔN of the predetermined rotational speed N is used, the error E as the estimated rotational speed N0 is reduced. In particular, the calculation of the change speed ΔN of the rotation speed N is likely to cause an error, but this embodiment does not need to use the change speed ΔN of the rotation speed N, and is advantageous in that the error due to the change speed ΔN is originally small.

  Further, according to the present embodiment, the predetermined rotational speed N includes the previously calculated estimated rotational speed N0 and two types of rotational speeds (for example, 1/4 rotational average rotational speed N4 and 1 rotational average rotational speed N16). Therefore, for example, the coefficients α and α ′ of the external component are appropriately set.

  Further, according to the present embodiment, the predetermined rotation speed N is equal to the estimated rotation speed N0 of two types of average rotation speed (for example, 1/4 rotation average rotation speed N4 and one rotation average rotation speed N16). Therefore, the average rotation speed with a relatively small response delay is used, for example, and the error E as the estimated rotation speed N0 is used. Is relatively small. In addition, when the estimated rotational speed N0 to be calculated is used, a recursive calculation results in a large calculation load. On the other hand, if the average rotational speed is used, such a recursive calculation can be avoided.

  Further, according to the present embodiment, the estimated rotational speed N0 is limited to be a numerical value within the range that can be taken as the actual rotational speed N, and therefore, for example, the estimated rotational speed N0 is a numerical value that cannot physically exist. Is avoided.

Further, according to the present embodiment, the sensor 60 is provided at a position facing a plurality of detected bodies (teeth 72, recesses 74) arranged along the circumferential direction on the radially outer peripheral portion of the rotating body 70. because it generates an interrupt signal S _PU based on sequentially relative teeth 72 (tooth a~ tooth p) and recesses 74 (the recess a~ recess p) in accordance with the rotation of the rotating body 70, for example, an interrupt signal _SPU is generated appropriately.

  Next, another embodiment of the present invention will be described. In the following description, parts common to the embodiments are denoted by the same reference numerals and description thereof is omitted.

  In the above-described embodiment, two types of average rotation speeds (for example, 1/4 rotation average rotation speed N4 and one rotation average rotation speed N16) are actually rotated in order to improve both detection accuracy and detection response. The estimated rotational speed N0 was calculated by placing it on the same time axis as the speed N. This embodiment also proposes an embodiment in which the detection response is improved by using a rotational speed that is delayed in time but relatively excellent in detection accuracy.

  Specifically, the electronic control unit 80 according to the present embodiment, for example, in order to achieve both improvement in detection accuracy and improvement in detection response when detecting the rotational speed N of the rotating body 70, for example, a signal for each latest tooth. To achieve both detection responsiveness when the interval tN [i] is directly converted to the rotational speed N and detection accuracy when calculating the average rotational speed using a plurality of consecutive signal intervals tN [i]. In addition, the estimated rotational speed N0 is calculated based on one rotational speed N calculated in the past and the change speed ΔN of the single rotational speed N. That is, the current actual rotation speed N is estimated from one rotation speed calculated in the past with relatively high detection accuracy using the gradient of the rotation speed. For example, one rotational speed N calculated in the past from a preset relationship between one rotational speed N calculated in the past and the estimated rotational speed N0 using the change speed ΔN of one rotational speed N calculated in the past as a parameter. And an estimated rotation speed N0 based on the change speed ΔN.

  One rotation speed N calculated in the past with relatively high detection accuracy is, for example, an estimated rotation speed N0 calculated last time and a 1/4 rotation average rotation calculated based on a signal interval tN [i] for a plurality of the past. For example, the average rotational speed such as the speed N4 and the average average rotational speed N16 is N16. However, similarly to the above-described embodiment, if the estimated rotation speed N0 is used, a recursive calculation may occur and the calculation load may be increased. Therefore, it is appropriate to use the average rotation speed. Furthermore, it is more appropriate to use an average rotation speed having a relatively small response delay with respect to the actual rotation speed N among the average rotation speeds. In the following description, in this embodiment, the 1/4 rotation average rotation speed N4 is exemplified as one rotation speed N calculated in the past.

  More specifically, the preset relationship between the one rotation speed N calculated in the past and the estimated rotation speed N0 is, for example, 1/4 rotation using the change speed ΔN4 of the 1/4 rotation average rotation speed N4 as a parameter. This is a relationship (estimated rotational speed map) as shown in FIG. 11 that is obtained experimentally (or by calculation) and set in advance between the average rotational speed N4 and the estimated rotational speed N0 itself. Alternatively, the preset relationship is, for example, the 1/4 rotation average rotation speed N4, the 1/4 rotation average rotation speed N4, and the estimated rotation speed N0 using the change speed ΔN4 of the 1/4 rotation average rotation speed N4 as a parameter. This is a relationship (estimated rotational speed difference map) as shown in FIG. 12, which is obtained experimentally (or by calculation) and set in advance with the difference D (= N4-N0). Alternatively, the preset relationship is, for example, the 1/4 rotation average rotation speed N4, the 1/4 rotation average rotation speed N4, and the estimated rotation speed N0 using the change speed ΔN4 of the 1/4 rotation average rotation speed N4 as a parameter. Is a relationship (estimated rotation speed rotation ratio map) as shown in FIG. 13 that is obtained experimentally (or by calculation) in advance and set with the rotation ratio R (= N0 / N4).

Returning to FIG. 8, the average rotation speed calculation means 84 replaces the above-described embodiment with, for example, the past four signal intervals (tN [0]) including the latest signal interval tN [0] based on the equation (1). + TN [1] + tN [2] + tN [3]), the 1/4 rotation average rotational speed N4 is sequentially calculated for each latest interrupt signal _SPU new. Further, the average rotation speed calculation means 84, for example, based on the previously calculated 1/4 rotation average rotation speed N4 and the 1/4 rotation average rotation speed N4 calculated this time, the change speed ΔN4 of the 1/4 rotation average rotation speed N4. Is calculated.

  The estimated rotational speed calculating means 86 is replaced with the 1/4 rotational average rotational speed N4 calculated by the average rotational speed calculating means 84 from the estimated rotational speed map as shown in FIG. An estimated rotational speed N0 is calculated based on the speed ΔN4.

  Alternatively, the estimated rotation speed calculation means 86 is based on the 1/4 rotation average rotation speed N4 calculated by the average rotation speed calculation means 84 and its change speed ΔN4, for example, from an estimated rotation speed difference map as shown in FIG. The difference D (= N4−N0) is calculated. Then, the estimated rotation speed calculation means 86 calculates an estimated rotation speed N0 (= N4-D) from the difference D based on the 1/4 rotation average rotation speed N4.

  Alternatively, the estimated rotational speed calculation means 86 may, for example, obtain a 1/4 rotational average rotational speed N4 calculated by the average rotational speed calculation means 84 and its change speed ΔN4 from an estimated rotational speed rotational ratio map as shown in FIG. Based on this, the rotation ratio R (= N0 / N4) is calculated. Then, the estimated rotation speed calculation means 86 calculates an estimated rotation speed N0 (= R × N4) from the rotation ratio R based on the 1/4 rotation average rotation speed N4.

  FIG. 14 is a flowchart for explaining a control operation for achieving both improvement in detection accuracy and improvement in detection response when detecting a main part of the control operation of the electronic control unit 80, that is, the rotational speed N of the rotating body 70. For example, it is repeatedly executed with an extremely short cycle time of about several milliseconds to several tens of milliseconds. The flowchart of FIG. 14 is another embodiment corresponding to the flowchart of FIG.

In FIG. 14, first, in step SB10 corresponding to the signal interval calculating means 82, for example, the latest signal interval tN [0] corresponding to the latest interrupt signal _SPU new generated with the rotation of the rotating body 70 is obtained. The latest interrupt signal _SPU new is calculated sequentially for each newest tooth (the latest tooth), and the calculation result is stored in the RAM. At this time, tN [0], signal interval tN [1], signal interval tN [2]... Calculated previously are tN [1], signal interval tN [2], signal interval tN [3]. To be rewritten. Next, in the SB 20 corresponding to the average rotation speed calculation means 84, for example, the 1/4 rotation average rotation speed N4 is calculated based on the past four signal intervals including the latest signal interval tN [0] from the equation (1). It is calculated sequentially for each latest interrupt signal _SPU new. Further, for example, based on the previously calculated 1/4 rotation average rotation speed N4 and the currently calculated 1/4 rotation average rotation speed N4, the change speed ΔN4 of the 1/4 rotation average rotation speed N4 is the latest interrupt signal S _PU. Calculated sequentially for each new. Next, in the SB 30 corresponding to the estimated rotation speed calculation means 86 and the estimated value guard means 88, for example, from the estimated rotation speed map as shown in FIG. 11, the 1/4 rotation average rotation speed N4 calculated in the SB 20 and its estimated rotational speed N0 based on the change rate ΔN4 is sequentially calculated for each date of the interrupt signal S _PU new new. Alternatively, for example, a difference D (= N4−N0) is calculated from an estimated rotation speed difference map as shown in FIG. 12 based on the 1/4 rotation average rotation speed N4 calculated in SB20 and the change speed ΔN4. Based on the difference D, the estimated rotational speed N0 (= N4-D) is sequentially calculated for each latest interrupt signal _SPU new based on the 1/4 rotation average rotational speed N4. Alternatively, for example, from the estimated rotation speed rotation ratio map as shown in FIG. 13, the rotation ratio R (= N0 / N4) based on the 1/4 rotation average rotation speed N4 calculated in SB20 and its change speed ΔN4. From the rotation ratio R, the estimated rotation speed N0 (= R × N4) is sequentially calculated for each latest interrupt signal _SPU new based on the 1/4 rotation average rotation speed N4. Note that the estimated rotational speed N0 calculated here is limited (guarded) so as to be a numerical value in a physically possible range as the actual rotational speed N. Accordingly, as in the above-described embodiment, for example, when the turbine rotation speed N _T (input rotation speed N _IN ) is changed, a 1/4 rotation average rotation speed N4 (broken line) calculated by a mere moving average. In addition, the estimated rotational speed N0 (solid line) with high responsiveness and accuracy is calculated as compared with the average rotational speed N16 (two-dot chain line). In particular, the detection accuracy in the low rotation speed region is improved, and control using this low rotation speed region is improved (see FIG. 10).

  As described above, according to this embodiment, the estimated rotation speed N0 is calculated based on one rotation speed calculated in the past and the change speed of the one rotation speed. The average rotational speed with good detection accuracy compared to the estimated rotational speed N0 calculated from tN [i] and the estimated rotational speed N0 calculated last time with good detection accuracy calculated using the average rotational speed are used. The estimated rotational speed N0 calculated automatically also has a relatively good detection accuracy. In addition, since the rotational speed at the time delayed in the time direction is calculated as the rotational speed at the same time as the actual rotational speed N, the finally calculated estimated rotational speed N0 is compared with the actual rotational speed N at the present time. Thus, there is no response delay. Therefore, when detecting the rotational speed N of the rotating body 70, it is possible to achieve both improvement in detection accuracy and improvement in detection response. Further, this embodiment has an advantage that only one rotation speed calculated in the past is required.

  Thereby, it becomes easy to grasp a physical phenomenon correctly. For example, the change in the rotational speed N detected by the rotating body 70 can be made closer to the actual change. Further, when comparing the rotational speeds of two or more different rotating bodies, for example, for detecting the operating state of the in-vehicle device such as detecting the speed change state of the speed change mechanism unit 16, There is no need to consider detection delay. Further, it is possible to advance the detection of the rotational speed N of the rotating body 70, and it is possible to accelerate the control starting point in the control operation of each device based on the detected rotational speed of each part. For example, inertia phase determination in the shifting process of the transmission mechanism 16, control to vary the phase of the camshaft according to the state of the engine 12, and coupling between the engine 12 and a power transmission device (for example, transaxle, transmission) when the vehicle is stopped Controls executed at a relatively low rotational speed, such as a neutral control for separating and a lock-up slip control for starting to lock the lock-up clutch from the start, can be accurately executed in accordance with changes in the actual rotational speed N. It becomes possible.

  Further, according to the present embodiment, from a preset relationship between one rotational speed N calculated in the past and the estimated rotational speed N0 using the change speed ΔN of one rotational speed N calculated in the past as a parameter, Since the estimated rotation speed N0 is calculated based on the calculated one rotation speed N and the change speed ΔN, for example, based on one rotation speed N calculated in the past and the change speed ΔN of the one rotation speed N. The estimated rotational speed N0 is calculated appropriately. That is, since one rotation speed calculated in the past is a value delayed from the actual rotation speed N, a value considering the delay time is calculated as the estimated rotation speed N0, so that the detection accuracy is improved and the detection response is improved. Estimated rotational speed N0 that can achieve both improvement is calculated.

  Further, according to the present embodiment, the one rotational speed N is, for example, the average rotational speed (1/4) calculated based on the estimated rotational speed N0 calculated last time and the signal intervals tN [i] for the past plural numbers. Therefore, the estimated rotation speed N0 is appropriately calculated. In particular, by using an average rotation speed that is relatively close in time to the estimated rotation speed N0 as the one rotation speed (for example, 1/4 rotation average rotation speed N4), for example, an average having a relatively small response delay. The rotational speed is used, and the error E as the estimated rotational speed N0 is made relatively small. In addition, when the estimated rotational speed N0 to be calculated is used, a recursive calculation results in a large calculation load. On the other hand, if the average rotational speed is used, such a recursive calculation can be avoided.

  Further, according to the present embodiment, the preset relationship is, for example, a direct relationship (estimated rotation) as shown in FIG. 11 between the 1/4 rotation average rotation speed N4 and the estimated rotation speed N0. Speed map). Alternatively, the preset relationship is set in advance, for example, a 1/4 rotation average rotation speed N4 and a difference D (= N4−N0) between the 1/4 rotation average rotation speed N4 and the estimated rotation speed N0. It is a relationship (estimated rotational speed difference map) as shown in FIG. Alternatively, the preset relationship is set in advance, for example, a 1/4 rotation average rotation speed N4 and a rotation ratio R (= N0 / N4) between the 1/4 rotation average rotation speed N4 and the estimated rotation speed N0. FIG. 13 is a relationship (estimated rotation speed rotation ratio map) as shown in FIG. In this way, for example, the estimated rotational speed N0 is more reliably calculated based on one rotational speed N calculated in the past and the change speed ΔN of the single rotational speed N.

  Further, according to the present embodiment, the estimated rotational speed N0 is limited to be a numerical value within the range that can be taken as the actual rotational speed N, and therefore, for example, the estimated rotational speed N0 is a numerical value that cannot physically exist. Is avoided.

  As mentioned above, although the Example of this invention was described in detail based on drawing, this invention is applied also in another aspect.

For example, in the illustrated embodiment, each signal interval tN [i] is was the spacing between each interrupt signal S _PU corresponding to the rise of the pulse signal P, other periods is not limited to this (the interval) The signal interval may be tN [i]. For example, each signal interval tN [i] may be an interval between interrupt signals corresponding to the falling edge of the pulse signal P, or each pulse signal Pa to Pp shown in FIG. Hi) or low (Lo) period, or the pulse signals Pa to Pp shown in FIG. 2 (c) are off (OFF). Also good. In short, the signal interval tN [i] corresponding to the rotational speed N of the rotating body 70 on a one-to-one basis may be used. Therefore, the pulse signals Pa to Pp having a constant ON period as shown in FIG. 2C do not need to be rectangular pulses. Further, a pulse signal P is generated based on a signal (voltage) generated when a detected object such as the tooth 72 passes through the sensor 60, and a signal corresponding to the rotational speed N of the rotating body 70 on a one-to-one basis. The present invention can be applied as long as the interval tN [i] can be calculated.

  In the above-described first embodiment, the one-rotation average rotation speed N16 and the 1 / 4-rotation average rotation speed N4 are exemplified as the two types of average rotation speeds. However, the present invention is not necessarily limited to such a form. obtain. In short, it is only necessary to use two types of average rotation speeds having different numbers of histories of the signal interval tN [i].

  Further, in the above-described second embodiment, the average rotation speed is exemplified by the one-rotation average rotation speed N16 and the quarter-rotation average rotation speed N4. However, the present invention is not necessarily limited to such a form. In short, an average rotation speed calculated based on the signal intervals tN [i] for a plurality of past may be used.

Further, in the above-described embodiment, the pulse signal P is generated corresponding to each tooth of the tooth 72 in a one-to-one manner. However, the present invention is not limited to such a form. In this case, a mode in which the pulse signal P is generated every several teeth instead of every tooth is also conceivable. The point is that the interrupt signal _SPU may be generated at every predetermined rotation angle of the rotating body 70.

  Further, in the above-described embodiment, the teeth 72 are exemplified as the plurality of detected bodies arranged along the circumferential direction on the outer peripheral portion in the radial direction of the rotating body 70. However, the embodiment is not necessarily limited to the form having such teeth 72. The present invention can be applied at least. For example, it is not a tooth 72 formed by punching on the rotating body 70 but a protrusion formed by forming a plurality of grooves arranged along the circumferential direction on the radially outer peripheral portion of the rotating body 70 by processing. May be. Further, the N pole and the S pole may be alternately arranged along the circumferential direction on the radially outer peripheral portion of the rotating body 70. In such a case, the protrusions and the N and S poles function as a plurality of detection objects. Further, although the number of teeth 72 is 16, the number of teeth is merely an example for explaining the rotational speed calculation processing of the present invention, and various modes are possible.

  In the above-described embodiment, the sensor 60 detects the rotational speed N of the rotating member 50 by detecting the rotational speed N of the rotating body 70. For example, a detected body (for example, a tooth) is provided. It may be provided at a position facing the rotating member 50 and directly detect the rotational speed N of the rotating member 50. In this case, the rotating member 50 itself corresponds to the rotating body.

  The above description is only an embodiment, and the present invention can be implemented in variously modified and improved forms based on the knowledge of those skilled in the art.

60: Sensor (rotation sensor, vehicle rotational speed detection device)
70: Rotating body 72, (ap): Teeth (detected body)
74, (ap): recessed portion (detected body)
80: Electronic control device (vehicle rotation speed detection device)
S _PU : Interrupt signal (interrupt signal)
tN [i]: Signal interval

Claims (12)

A vehicle rotation speed detection device that includes a rotation sensor that generates an interruption signal for each predetermined rotation angle of a rotation body, and that detects the rotation speed of the rotation body based on a signal interval of the interruption signal,
Two types of average rotation speeds having different numbers of history of the signal intervals are calculated, and the two types of average rotation speeds are placed on the same time axis as the actual rotation speeds, thereby calculating an estimated value of the actual rotation speeds. Is,
The two types of average rotation speeds are average rotation speeds calculated based on different numbers of signal intervals for the past plurality of signals including the latest signal interval,
An external division that externally divides the two types of average rotational speeds based on respective time-direction delays that occur between the time point when the estimated value of the actual rotational speed is calculated and the time point corresponding to the two types of average rotational speeds An estimated value of the actual rotational speed is calculated as a point . The coefficient of the external component based on each time direction delay for calculating the estimated value of the actual rotational speed is set to the predetermined rotational speed after fixing the changing speed of the predetermined rotational speed to a constant value. The vehicle speed detection device according to claim 1 , wherein a value that changes in accordance with the value is used. The vehicle rotational speed detection device according to claim 2 , wherein the predetermined rotational speed uses one of an estimated value of the actual rotational speed calculated last time and the two types of average rotational speeds. The vehicle-use vehicle according to claim 2 , wherein the predetermined rotation speed is an average rotation speed that is temporally closer to the estimated value of the actual rotation speed among the two types of average rotation speeds. Rotational speed detection device. Estimated value of the actual rotational speed, the rotational vehicle according to any one of claims 1 to 4, characterized in that it is limited to be a number in the range of possible as the rotational speed of said actual Speed detection device. A vehicle rotation speed detection device that includes a rotation sensor that generates an interruption signal for each predetermined rotation angle of a rotation body, and that detects the rotation speed of the rotation body based on a signal interval of the interruption signal,
An estimated value of the actual rotation speed is calculated based on one rotation speed calculated in the past and the change speed of the one rotation speed ,
Based on a preset relationship between the one rotation speed and the estimated value of the actual rotation speed, using the change speed of the one rotation speed as a parameter, the one rotation speed and the change speed of the one rotation speed are changed. An estimation value of an actual rotation speed is calculated based on the vehicle rotation speed detection device. The vehicle rotational speed detection device according to claim 6 , wherein the preset relationship is a direct relationship between the one rotational speed and an estimated value of the actual rotational speed. 7. The vehicle-use vehicle according to claim 6 , wherein the preset relationship is a relationship between the one rotation speed and a difference between the one rotation speed and an estimated value of the actual rotation speed. Rotational speed detection device. 7. The vehicle according to claim 6 , wherein the preset relationship is a relationship between the one rotation speed and a rotation ratio between the one rotation speed and an estimated value of the actual rotation speed. Rotational speed detection device. It said one of the rotational speed, the actual rotational speed of the estimated value previously calculated, or any claim 6 to 9, characterized in that the average rotational speed calculated based on the signal interval of the past several minutes The vehicle rotational speed detection device according to claim 1. The vehicle rotation according to any one of claims 6 to 10 , wherein the estimated value of the actual rotation speed is limited to a value within a range that can be taken as the actual rotation speed. Speed detection device. The rotation sensor is provided at a position facing a plurality of detected bodies arranged in a circumferential direction on a radially outer peripheral portion of the rotating body, and sequentially with the detected body as the rotating body rotates. The vehicle rotation speed detection device according to any one of claims 1 to 11 , wherein an interrupt signal is generated on the basis of the opposite.

Images